Clinically-related collaborative studies were conducted in several areas, using the vestibular autorotatin test (VAT). A large series of normal subjects and selected patients were tested on the same day with both the VAT and a sum-of-sines rotational chair protocol. The goal is to test a series of hypotheses concerning the relative similarities or differences in VAT vs chair results over a broad-band frequency range from 0.02 to 2 Hz. Results of testing normal pediatric subjects (ages 3-18) with the VAT were completed. These results are being extended for further collaborative studies on diagnostic screening of pediatric patients. A new project was begun using neural network classification of Danish air force flight candidates. Sixty student pilots in training for flight qualification of F16 fighter jets were tested with the VAT before beginning training. VAT parameters were used in a back-propagation method of training a neural network to classify which of the students would successfully qualify for F16 flight certification. Results were shown to be accurately predicted with greater than 90 percent accuracy. These methods will be adapted in the future to classification of selected clinical categories for predicting the success of rehabilitation of patients with vestibular disorders.